def test_decision_function(self):
from lale.lib.imblearn import SMOTE
from lale.operators import make_pipeline
from lale.lib.sklearn import RandomForestClassifier
smote = SMOTE(operator=make_pipeline(RandomForestClassifier()))
trained = smote.fit(self.X_train, self.y_train)
trained.predict(self.X_test)
with self.assertRaises(AttributeError):
trained.decision_function(self.X_test)
def test_random_forest_classifier(self):
ranges, dists = RandomForestClassifier.get_param_ranges()
expected_ranges = {
"n_estimators": (10, 100, 100),
"criterion": ["entropy", "gini"],
"max_depth": (3, 5, None),
"min_samples_split": (2, 5, 2),
"min_samples_leaf": (1, 5, 1),
"max_features": (0.01, 1.0, 0.5),
}
self.maxDiff = None
self.assertEqual(ranges, expected_ranges)
def test_max_samples(self):
with self.assertRaisesRegex(jsonschema.ValidationError,
"argument 'max_samples' was unexpected"):
_ = RandomForestClassifier(max_samples=0.01)
def test_ccp_alpha(self):
with self.assertRaisesRegex(jsonschema.ValidationError,
"argument 'ccp_alpha' was unexpected"):
_ = RandomForestClassifier(ccp_alpha=0.01)
def test_n_estimators(self):
default = RandomForestClassifier.hyperparam_defaults()["n_estimators"]
self.assertEqual(default, 10)
def test_with_defaults(self):
trainable = RandomForestClassifier()
trained = trainable.fit(self.train_X, self.train_y)
_ = trained.predict(self.test_X)
def test_max_samples(self):
trainable = RandomForestClassifier(max_samples=0.01)
trained = trainable.fit(self.train_X, self.train_y)
predicted = trained.predict(self.test_X)
def test_ccp_alpha(self):
trainable = RandomForestClassifier(ccp_alpha=0.01)
trained = trainable.fit(self.train_X, self.train_y)
predicted = trained.predict(self.test_X)
